The overall aim of this research is to identify targets of CpG island alterations associated with human colon tumorigenesis using Restriction Landmark Genomic Scanning (RLGS). Using Notl or other restriction enzymes that have GC rich recognition sequences as the restriction landmark, RLGS targets gene rich CpG island regions and primarily detects alterations in DNA methylation status although DNA deletions and amplifications can also be detected. RLGS loci exhibiting frequent alterations associated with human colon tumorigenesis will be cloned using in silico based, virtual RLGS computational methods. The confirmation of identity between the "virtual" RLGS sequence and the "real" RLGS sequence will be accomplished by PCR amplification of the RLGS spot DNA using primers derived from the virtual sequence. Array Comparative Genomic Hybridization (CGH) will be performed on the same tumors analyzed by RLGS to integrate epigenetic and genetic aberrations associated with human colon cancer. This will allow us to establish whether there is a class of tumors with high epigenetic instability that can be distinguished from those with high genetic instability and whether there is cooperative interaction between epigenetic and genetic aberrations. The methylation status of selected RLGS loci will be evaluated by methylation sensitive PCR and bisulfite sequencing to determine the boundaries and density of DNA methylation. Quantitative expression assays will be used to assess the expression of putative target genes in primary tumors to determine whether the CpG island alteration was associated with a change in gene expression. The identification of novel targets of CpG island alteration may uncover new pathways of tumorigenesis and further our understanding of how epigenetic alterations impact on the malignant phenotype.